Scribble-based colorization for creating smooth-shaded vector graphics

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

9 Scopus Citations
View graph of relations


Related Research Unit(s)


Original languageEnglish
Pages (from-to)73-81
Journal / PublicationComputers and Graphics (Pergamon)
Online published21 Apr 2019
Publication statusPublished - Jun 2019


This paper proposes a novel colorizaiton tool for intuitively creating smooth-shaded vector graphics. Our technique takes advantage of diffusion curves, a powerful vector graphics representation. Rather than specifying colors along curves as done in previous works, we allow a user to intuitively paint in regions of arbitrary line drawings. Once the user scribbles on drawings, our algorithm automatically estimates the colors along the curves of drawings, resulting in smooth color regions as close as possible to the user specification. Different from the previous color estimation techniques for image vectorization, A new diffusion curve colorization algorithm is proposed for fitting sparse colors of input scribbles. Our approach is fast and provides instant feedback to the user. We have tested our system on a variety of line drawings with varying shape complexity, and shown that our technique can produce visually pleasing smooth-shaded images intuitively and effectively.

Research Area(s)

  • Colorization, Diffusion curves, Scribble-based UI, Vector graphics